Unsupervised Neighborhood Propagation Kernel Layers for Semi-supervised Node Classification

TL;DR

This paper introduces a novel deep graph kernel method for semi-supervised node classification that leverages unsupervised neighborhood propagation and kernel machines to improve performance with limited labels.

Contribution

It proposes a new architecture combining unsupervised kernel layers with semi-supervised classification, along with an efficient training scheme.

Findings

Effective in semi-supervised settings with few labels

Outperforms existing methods on benchmark datasets

Demonstrates robustness to limited labeled data

Abstract

We present a deep Graph Convolutional Kernel Machine (GCKM) for semi-supervised node classification in graphs. The method is built of two main types of blocks: (i) We introduce unsupervised kernel machine layers propagating the node features in a one-hop neighborhood, using implicit node feature mappings. (ii) We specify a semi-supervised classification kernel machine through the lens of the Fenchel-Young inequality. We derive an effective initialization scheme and efficient end-to-end training algorithm in the dual variables for the full architecture. The main idea underlying GCKM is that, because of the unsupervised core, the final model can achieve higher performance in semi-supervised node classification when few labels are available for training. Experimental results demonstrate the effectiveness of the proposed framework.